Method Drift›Retrieval-augmented generation
RAG-Star
RAG-Star: Enhancing Deliberative Reasoning with Retrieval Augmented Verification and RefinementRetrieval-augmented generation · first seen Dec 17, 2024
superseded — cited as a baseline and beaten by newer methods
1 papers critique it · 1 beat it on benchmarks
What papers say
Verbatim critique sentences, each from a paper that cites RAG-Star as a baseline.
“RAG-Star~jiang2024rag uses a fixed search tree for token-level retrieval control, but lacks dynamic pruning and does not support concurrent reasoning paths.”
— MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
Beaten on benchmarks
Head-to-head results where a newer method reports beating RAG-Star. Values are copied from the source paper's tables — verify against the cited paper.
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats RAG-Star · CWQA [Qwen2.5-7B]
61.4 vs 58.1
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats RAG-Star · GPQA [Qwen2.5-7B]
64.6 vs 57.9
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats RAG-Star · FMT [Qwen2.5-7B]
68.3 vs 67.4
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats RAG-Star · CWQA [Llama 3.1-8B]
67.3 vs 60.1
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats RAG-Star · GPQA [Llama 3.1-8B]
71.3 vs 66.3
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats RAG-Star · FMT [Llama 3.1-8B]
73.8 vs 69.8
Newer alternatives
Recent methods in the same sub-problem, not yet superseded in the knowledge base.